Meaning-based advertising and document relevance determination

• US 6,816,857 B1
• Filed: 01/28/2000
• Issued: 11/09/2004
• Est. Priority Date: 11/01/1999
• Status: Expired due to Term

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Applied Semantics, Inc. (Alphabet Inc.)
• Inventors: Weissman, Adam J., Elbaz, Gilad Israel
• Primary Examiner(s): Wassum, Luke S
• Application Number:
  US09/493,701
• Time in Patent Office: 1,747 Days
• Field of Search: 707/3-5, 707/100, 707/104.1
• US Class Current: 1/1
• CPC Class Codes: G06F 16/243 Natural language query formulation
• CPC Class Codes: G06F 16/24522 Translation of natural language queries to structured queries
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06Q 30/0256 User search
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

The present invention is directed to a system in which a semantic space is searched in order to determine the semantic distance between two locations. A further aspect of the present invention provides a system in which a portion of semantic space is purchased and associated with a target data set element which is returned in response to a search input. The semantic space is created by a lexicon of concepts and relations between concepts. An input is associated with a location in the semantic space. Similarly, each data element in the target data set being searched is associated with a location in the semantic space. Searching is accomplished by determining a semantic distance between the first and second location in semantic space, wherein this distance represents their closeness in meaning and where the cost for retrieval of target data elements is based on this distance.

Claims

1. A method comprising:

organizing concepts according to their meaning into a lexicon of predefined known relationships between the concepts, said lexicon defining elements of a semantic space;

receiving a first input and, based on interpretation of potential meanings, associating that input with a first set of concepts from said lexicon, said first input representing a first location in the semantic space;

assigning a probability factor for each concept in the first set of concepts for the input received;

receiving a second input associated with a second set of concepts from said lexicon, said second input representing a second location in the semantic space;

determining a semantic distance from the first location to the second location by combination of the semantic distance between each concept in the first set of concepts and each concept in the second set of concepts; and

presenting results of a search conducted on the second set of concepts for data associated with the second set of concepts close in meaning to the first input based on the determined semantic distances and probability factors associated with the concept in the first set of concepts.

2. A method comprising:

creating a lexicon of predefined known relationships between the concepts that defines elements of a semantic space;

receiving an input and, based on interpretation of potential meanings, associating that input with a first set of concepts from said lexicon and representing a first location in said semantic space;

assigning a probability factor for each concept in the first set of concepts for the input received;

maintaining a target data set, wherein the target data is associated with a second set of concepts from said lexicon that presents a second location in the semantic space;

determining a semantic distance from the first location in the semantic space to the second location in the semantic space by combination of the relative closeness in meaning between each concept in the first set of concepts and each concept in the second set of concepts; and

presenting results of a search conducted on the target data set for target data close in meaning to the input based on the determined semantic distances and probability factors associated with the concept in the first set of concepts.

3. A method according to claim 2 wherein the input is a data item and the associated set of concepts represents at least one of the meanings of said data item and important concepts relevant to the data item.

4. A method according to claim 3 wherein said data item is text.

5. A method according to claim 4 wherein said text is derived from the conversion of a printed text to electronic form.

6. A method according to claim 4 wherein said text is derived from audio data.

7. A method according to claim 4 wherein said text is derived from video data.

8. A method according to claim 4 wherein said text is used to label an entity.

9. A method according to claim 8 wherein said labeled entity is one from the group of an image, video, sound file or document.

10. A method according to claim 8 wherein said labeled entity is a person and where the labeling represents data about the person such as his interests or geographical location.

11. A method according to claim 4 wherein said text is a user query.

12. A method according to claim 4 wherein said text is a domain name or a full URL.

13. A method according to claim 4 wherein said text is a document.

14. A method according to claim 4 wherein said text is web content.

15. A method according to claim 4 wherein said text is an electronic communication.

16. A method according to claim 3 wherein said data item is one from the group of an image, video, sound file or document.

17. A method according to claim 3 wherein said set of concepts is associated with a person.

18. A method according to claim 3 wherein said data item is one from the group of an advertisement, a product or service or a category.

19. A method according to claim 3 wherein said set of concepts associated with said data item is predetermined.

20. A method according to claim 19 wherein said set of concepts associated with said data item is specified by a user.

21. A method according to claim 19 further comprising enabling a user to select at least one meaning from the set of possible meanings for said data item in order to provide the correct interpretation for establishing a set of concepts representing the meaning of the data item.

22. A method according to claim 2 wherein the target data is a data item and the associated set of concepts represents at least one of the meaning of said data item and important concepts relevant to the data item.

23. A method according to claim 22 wherein said data item is text.

24. A method according to claim 23 wherein said text is derived from the conversion of a printed text to electronic form.

25. A method according to claim 23 wherein said text is derived from audio data.

26. A method according to claim 23 wherein said text is derived from video data.

27. A method according to claim 23 wherein said text is used to label an entity.

28. A method according to claim 27 wherein said labeled entity is one from the group of an image, video, sound file or document.

29. A method according to claim 27 wherein said labeled entity is a person and where the labeling represents data about the person such as his interests or geographical location.

30. A method according to claim 27 wherein said labeled entity is one from the group of an advertisement, a product or service, or a category.

31. A method according to claim 23 wherein said text is a user query.

32. A method according to claim 23 wherein said text is a domain name or a full URL.

33. A method according to claim 23 wherein said text is a document.

34. A method according to claim 23 wherein said text is web content.

35. A method according to claim 23 wherein said text is an electronic communication.

36. A method according to claim 22 wherein said data item is one from the group of an image, video, sound file or document.

37. A method according to claim 22 wherein said set of concepts is associated with a person.

38. A method according to claim 22 wherein said data item is one from the group of an advertisement, a product or service or a category.

39. A method according to claim 22 wherein said set of concepts associated with said data item is predetermined.

40. A method according to claim 39 wherein said set of concepts associated with said data item is specified by a user.

41. A method according to claim 39 further comprising enabling a user to select at least one meaning from the set of possible meanings for said data item in order to provide the correct interpretation for establishing a set of concepts representing the meaning of the data item.

42. A method according to claim 2 wherein said second location is assigned a monetary value.

43. A method according to claim 42 wherein the price of being included in a target data set is determined by the monetary value of a set of concepts it is associated with.

44. A method according to claim 42 wherein the monetary value of a set of concepts is based on the cost of its sub-space in the semantic space.

45. A method according to claim 42 wherein the price of retrieving target data is dynamically calculated in response to an input query, the price of returning the target data in the result increasing with the relevance of its associated set of concepts to the query.